Effects of combinatorial treatment with pituitary adenylate cyclase activating peptide and human mesenchymal stem cells on spinal cord tissue repair

Abstract

The aim of this study is to understand if human mesenchymal stem cells (hMSCs) and neuropeptide pituitary adenylate cyclase-activating polypeptide (PACAP) have synergistic protective effect that promotes functional recovery in rats with severe spinal cord injury (SCI). To evaluate the effect of delayed combinatorial therapy of PACAP and hMSCs on spinal cord tissue repair, we used the immortalized hMSCs that retain their potential of neuronal differentiation under the stimulation of neurogenic factors and possess the properties for the production of several growth factors beneficial for neural cell survival. The results indicated that delayed treatment with PACAP and hMSCs at day 7 post SCI increased the remaining neuronal fibers in the injured spinal cord, leading to better locomotor functional recovery in SCI rats when compared to treatment only with PACAP or hMSCs. Western blotting also showed that the levels of antioxidant enzymes, Mn-superoxide dismutase (MnSOD) and peroxiredoxin-1/6 (Prx-1 and Prx-6), were increased at the lesion center 1 week after the delayed treatment with the combinatorial therapy when compared to that observed in the vehicle-treated control. Furthermore, in vitro studies showed that co-culture with hMSCs in the presence of PACAP not only increased a subpopulation of microglia expressing galectin-3, but also enhanced the ability of astrocytes to uptake extracellular glutamate. In summary, our in vivo and in vitro studies reveal that delayed transplantation of hMSCs combined with PACAP provides trophic molecules to promote neuronal cell survival, which also foster beneficial microenvironment for endogenous glia to increase their neuroprotective effect on the repair of injured spinal cord tissue.

Figure 1. Neuronal-like transformation of immortalized hMSCs after treatment with neurogenic factors.

The immortalized hMSCs were treated for 24 hours in serum-free ITS medium with several neurogenic factors, dbcAMP (0.1 mM), PACAP (20 ng/ml), β-mercaptoethanol (β-ME; 1 mM), and retinoic acid (RA; 0.1 µM). The cultures were subjected to immunofluorescence staining for α-internexin (A–F) or synapsin (G–L). Arrows and arrowheads indicate elongation and branching of the processes in hMSCs, respectively. Scale bar in A–L, 100 µm.

Figure 2. Secretion profile of growth factors and cytokines from immortalized hMSCs.

Immortalized hMSCs were cultured in DMEM/LG medium for 48 hours. The cultured media were collected and applied onto a human growth factors/cytokines antibody array membrane (left panel in A). The map of the array that is designed to detect 60 cytokines, chemokines, or growth factors is shown in A (right panel). The arrays were scanned, and the staining intensity of each spot was quantified by densitometry. The percentage of the intensity of the respective spots over that of the positive control (POS) on the same array was represented as the relative expression levels of each human growth factor (B) or cytokine (C). The data represent mean ± SEM of three independent experiments.

Figure 3. Antioxidant proteins and hindlimb locomotion increased by delayed combinatorial therapy of PACAP and hMSCs.

(A). The diagram shows the location of the administration of hMSCs and PACAP into 1–2 mm rostral and caudal to the lesion center at day 7 post severe SCI. (B) Immunostaining for human nucleus (HuNu) was carried out to identify transplanted hMSCs (green) at day 7 after hMSC transplantation. Many hMSCs were present in the injected zone proximal to the lesion center of the spinal cord. Scale bar in B, 50 µm. (C). The injured spinal cords were collected at day 14 after SCI. The lesion center with the length of 4–5 mm was dissected from the injured spinal cord tissues, subjected to protein extraction. The expression levels of selected antioxidant proteins as indicated above were examined by western blotting. The same blot was stripped and reprobed with β-actin antibodies as internal loading control. (D). At day 31 after SCI, rats with severe SCI were subjected onto an open-field to evaluate their hindlimb locomotion using Basso Beattie Bresnahan (BBB) locomotor rating analysis. Data are presented as mean ± SEM. *p<0.05 versus vehicle, PACPA, or hMSCs.

Figure 4. Immunofluorescence staining for identifying the remaining neuronal fibers in the lesion center of the injured spinal cord.

After BBB open field scores were performed at day 31, spinal cord tissues were collected. Longitudinal tissue sections were subjected to immunofluorescence for CGRP (A–D), NF-200 (E–H), or GAP-43 (I–L). CGRP positive neuronal fibers were observed in the dorsal section of the spinal cord (D, arrows). Elongated neuronal bundles labeled with CGRP, NF-200, or GAP-43 (arrows in D, H, and L) were observed in the injured spinal cord tissue receiving delayed combinatorial treatment of PACAP and hMSCs. Few NF positive or GAP-43 positive neuronal fibers were seen in the injured spinal cord tissues treated with PACAP (arrows in F and open arrows in G). Fine NF positive fibers (arrow in G) or fragmented GAP-43 positive fibers (open arrows in K) can be detected in the lesion center of the spinal cord with delayed treatment by hMSCs. Arrowheads indicate the cell debris with immunostaining for CGRP (A), NF-200 (E), or GAP-43 (I). Scale bar, 100 µm.

Figure 5. Immunofluorescence of GFAP and Iba-I in the injured spinal cord tissues. The injured spinal cord tissue sections were collected at day 31 after SCI.

The longitudinal tissue sections were subjected to immunofluorescence for GFAP (A–D) and Iba-1 (E–H). GFAP positive hypertrophic cells (open arrows) were observed in the areas proximal to the lesion center of the injured spinal cord treated with vehicle (A), PACAP (B), or hMSCs (C), while GFAP positive stellated cells (arrowheads) were found at the periphery to the lesion center treated with hMSCs and PACAP (D). Amoeboid-shaped Iba-1 positive microglia/macrophages (arrowheads) were observed in the injured spinal cord proximal to the lesion center with combinatorial treatment by hMSCs and PACAP (H). Unipolar Iba-1 positive microglia (arrows in E) are present around the lesion center of the spinal cord treated with vehicle. Scale bars, 100 µm (A–D) and 50 µm (E–H).

Figure 6. Galectin-3 positive microglia increased by combinatorial treatment with PACAP and hMSCs.

(A). The injured spinal cord was collected at day 14 post SCI, which was sectioned and subjected to immunofluorescence staining for galectin-3 (red, arrowhead)/HuNu(green, arrow) and Iba-1 (red, arrowhead)/HuNu (green, arrow). The tissue sections were then incubated with DAPI for nuclear counterstain. (B). Microglia-hMSC co-cultures were treated with PACAP (100 nM) for 24 hours, and then subjected to immunofluorescence staining for galectin-3 (green) and HuNu (red). (C). A representative cytogram of microglia and hMSCs was shown in the left and middle panel, respectively. Accordingly, the cytogram of microglia-hMSC co-culture indicates two cell populations (right panel): microglia scattering in R1, and hMSCs appearing in R2. (D) Microglia were treated with PACAP (100 nM), or co-cultured with hMSCs in the absence or presence of PACAP (100 nM). 24 hours later, the relative levels of galectin-3 positive microglia were determined by FACS. The data represent as the relative level of galectin-3 positive microglia by determining the ratio of galectin-3 positive microglia being analyzed in the region of R1 (C) in comparison with that detected in the control culture. The images shown in B and C were taken from confocal microscopy. Data consist of means ± SEM of three independent experiments. Scale bar in A, 50 µm; in B, 40 µm. *p<0.05 versus control.

Figure 7. GLT-1 production and glutamate uptake ability of astrocytes promoted by combinatorial treatment with PACAP and hMSCs.

(A). Astrocytes were treated with PACAP (20 ng/ml), or indirectly co-cultured with hMSCs in the absence and presence of PACAP (20 ng/ml) for 48 hours. The cultures were subjected to Q-PCR for measurement of the mRNA levels of GLT-1 and GLAST. Data are presented as mean ± SEM and expressed as the ratio of GLT-1 (GLAST) mRNA levels compared to control. *p<0.05 versus control. (B). Astrocytes were indirectly co-cultured with hMSCs in the absence or presence of PACAP for 48 hours. Membrane proteins were then extracted for western blotting using anti-GLT-1 antibody. The same blot was reprobed with anti-P2X₇R antibody. P2X₇R levels are presented as a loading control. Relative intensity of GLT-1 levels normalized to P2X₇R was measured. Data are presented as mean ± SEM and expressed as a percentage of GLT-1 levels in the group with combinatorial treatment compared to that detected in the group co-cultured only with hMSCs. *p<0.05 versus the group only co-cultured with hMSCs. (C). Astrocytes were infected by lentivirus carrying lenti-GLT-1-GFP, and then co-cultured with hMSCs in the absence or presence of PACAP. Strong punctate fluorescence spots (arrowheads) were observed in the processes of astrocytes co-cultured with hMSCs in the presence of PACAP, when compared to that seen in astrocytes co-cultured only with hMSCs (arrows). Scale bar, 50 µm. (D). Astrocytes were treated with PACAP, or co-cultured with hMSCs in the absence or presence of PACAP. After 48 or 72 hours, the cultures were subjected to [³H]-L-glutamate uptake analysis as described in Materials and Methods. Data are presented as mean ± SEM and expressed as a ratio of the glutamate uptake in each treated group compared to that detected in the control group. *p<0.05 versus control.